https://doi.org/10.5194/nhess-2020-427 Preprint. Discussion started: 27 January 2021 c Author(s) 2021. CC BY 4.0 License.

1 Assessing local impacts of the A.D. 1700 Cascadia and using tree ring growth

2 histories: A case study in South Beach, , U.S.A.

3 Robert P. Dziak1, Bryan A. Black2, Yong Wei3, and Susan G. Merle4 4 1NOAA/Pacific Marine Environmental Laboratory, Newport, Oregon, 97365 U.S.A. 5 2Laboratory of Tree-Ring Research, University of Arizona, Tucson, Arizona, U.S.A. 6 3NOAA/Pacific Marine Environmental Laboratory, , , 98115 U.S.A. 7 4Cooperative Institute for Marine Resources Studies, , Newport, Oregon, 97366 U.S.A. 8 Correspondence to: Robert P. Dziak ([email protected])

9 Abstract. We present a spatially focused investigation of the disturbance history of an old-growth Douglass fir stand in South 10 Beach, Oregon for possible growth effects due to tsunami inundation caused by the A.D. 1700 11 earthquake. A high-resolution model of the 1700 tsunami run-up heights at South Beach, assuming an “L” sized earthquake, 12 is also presented to better estimate the inundation levels several kilometers inland at the old-growth site. This tsunami model 13 indicates the South Beach fir stand would have been subjected to local inundation depths from 0-10 m. Growth chronologies 14 collected from the fir stand shows several trees experienced significant growth reductions before, during and several years 15 after 1700, consistent with the tsunami inundation estimates. The +/- 1-3 year timing of the South Beach disturbances are also 16 consistent with disturbances previously observed at a Washington state coastal forest ~220 km to the north. Additional 17 comparison of the South Beach chronologies with regional chronologies across Oregon indicates the South Beach stand growth 18 was significantly and unusually lower in 1700. Moreover, the 1700 South Beach growth reductions were not the largest over 19 the 110-year tree chronology at this location. with other disturbances likely caused by other climate drivers (e.g. drought or 20 windstorms). Our study represents a first step in using tree growth history to ground-truth tsunami inundation models by 21 providing site specific physical evidence. 22 .

23 1. Introduction

24 Recent catastrophic tsunami inundation events along the Sumatra and Japanese coasts have shown can have 25 a devastating effect on coastal forests and overall coastal geomorphology [Kathiresan and Rajendran, 2005; Udo et 26 al., 2012; Lopez Caceres et al., 2018]. In addition to the tsunami, ground motion caused by the 27 can cause significant forest disturbance by toppling trees, damaging root systems, severing limbs and crowns, 28 inducing damaging , or altering the hydrology of a stand, among other potential effects [e.g. Shepphard and 29 Jacoby, 1989]. These disturbances appear in the tree-ring record of surviving trees as sudden suppression events 30 (when there is damage), or growth events in the case of reduced competition from adjacent damaged trees.

1

https://doi.org/10.5194/nhess-2020-427 Preprint. Discussion started: 27 January 2021 c Author(s) 2021. CC BY 4.0 License.

31 32 Here we present a spatially focused investigation of the disturbance history of an old-growth forest in South Beach, 33 Oregon (Figure 1). We also present a new, high resolution model of the 1700 tsunami run-up heights at South Beach 34 to better estimate the inundation levels at the site of the old-growth forest. Our goal is to use tree-growth to ground- 35 truth the tsunami impacts and inundation levels as well as for insights into the degree of shaking caused by the 1700 36 magnitude 9.0 Cascadia Subduction Zone earthquake [Satake et al., 2003; Witter et al.., 2011]. 37 38 Interestingly, direct evidence of seismic shaking (liquefaction, landslides, etc) from the 1700 megathrust earthquake 39 is relatively rare along the Oregon coast. This is thought to be due to the high rainfall and water erosion rates in the 40 which removes liquefaction evidence in coastal estuaries, and makes landslides in the coast range 41 difficult to identify [Yeats, 2004; LaHusen et al., 2020]. Models of shaking and ground motion along the Oregon 42 coast during the 1700 Cascadia earthquake indicate it should have been violent and widespread [WDNR, 2012], and 43 it seems plausible that evidence of this shaking might be recorded in the ring widths of trees along the coast. 44 45 Very little tree-ring work has been conducted along the Oregon coast; the vast majority of tree-ring research in the 46 Pacific Northwest has entailed climate reconstructions from high-elevation sites in the Cascade Mountains and 47 Olympic Peninsula where competitive effects are low. We sampled a mesic old-growth forest near the Pacific coast 48 where competitive effects are high. Significant disturbances from the 1700 earthquake and tsunami should 49 substantially alter radial growth patterns as some trees are damaged or killed and resources are redistributed to 50 survivors. Alternatively, the tsunami may cause physical damage to trees resulting in growth reductions. The goal of 51 this study is to investigate whether these disturbances are observable in the few remaining old-growth forests along 52 the coast of Oregon. Thus we chose a site where good inundation models exist, and there is significant public concern 53 about tsunami impacts because of the presence of a large population and municipal infrastructure. 54 55 2.0 Evidence for Megathrust and Tsunamis 56 57 On 26 January at 9:00PM in the year 1700 A.D., a large earthquake occurred on the Cascadia Subduction Zone, the 58 boundary between the Pacific and North American plates along the coasts of California, Oregon, Washington, and 59 British Columbia [Satake et al., 2003]. The earthquake created a tsunami with 10-12 m run-up heights that struck the 60 Pacific Northwest and propagated across the Pacific to Japan [Atwater, 1992; Satake, et al, 2003; Goldfinger et al., 61 2003]. The 1700 earthquake is estimated to have most likely been a moment magnitude (Mw) 9.0, with between 13- 62 21 m of coseismic slip on an offshore 1100 km long [Satake et al., 2003; Witter et al., 2011]. The 1700 earthquake 63 was preceded by an earthquake in 960 A.D. (740 yr interval) and another in 750 A.D. (210 yr interval), with three 64 additional subduction events before these that comprise a recent cluster of 6 megathrust events over the past 1500 yrs

2

https://doi.org/10.5194/nhess-2020-427 Preprint. Discussion started: 27 January 2021 c Author(s) 2021. CC BY 4.0 License.

65 [Atwater et al., 2003]. During the 1700 Cascadia earthquake, ground motion and peak ground acceleration (PGA), 66 are modeled to have ranged from ~0.5-1.2 g along the Oregon coast [WDNR, 2012]. Thus ground motion shaking 67 during the 1700 earthquake should have been violent and widespread. 68 69 The exact timing of the earthquake was estimated by calculating the travel time for an unexplained tsunami that struck 70 Japan on 26 January 1700 [Satake et al., 2003; Atwater, 2006]. Radiocarbon dating was used to show the earthquake 71 occurred in 1700, where the radiocarbon dates were derived from the remnants of many trees drowned by coincident 72 subsidence, and surviving trees recorded the earthquake’s date by anomalous changes in ring width or anatomy of 73 their annual rings [Atwater and Yamaguchi, 1991; Jacoby et al., 1997]. As a result of subsidence, some coastal forests 74 dropped below sea level and were flooded. Boles and root masses of these trees still remain and can be found from 75 northern Oregon to southern Washington. Aligning tree-ring growth patterns of the living trees with those of the 76 flooded, dead trees consistently showed that the last year of growth was 1699, indicating the earthquake occurred 77 between October 1699 and April 1700 [Yamaguchi et al., 1997]. This tree-ring and dating evidence for coastal 78 disturbance is indeed compelling, however the evidence was derived from trees along just 100 km of coastal southern 79 Washington-northern Oregon, or ~5% of the coastline expected to be affected by a Cascadia megathrust earthquake. 80 81 Additionally, a coastal-wide inventory of liquefaction features associated with the 1700 earthquake found no features 82 along the Oregon coast, despite numerous exposures of clean sand deposits that must be susceptible to liquefaction, 83 even at low levels of seismic shaking [Obermeier and Dickenson, 2000]. The locations for these field studies in 84 Oregon were also sites where evidence for great Holocene subduction earthquakes (in the form of crustal subsidence) 85 have been identified [Nelson et al., 1995]. The only liquefaction features identified to date (and thus direct evidence 86 of seismic shaking) were found along the 35-50 km east of the coast, and these indicate moderate 87 shaking intensity of 0.2-0.35 g [Obermeier and Dickenson, 2000]. 88 89 3.0 Model of A.D. 1700 Tsunami 90 91 As a first step in estimating tree disturbance in South Beach, we produced a model of tsunami inundation level and 92 expected flow speed for the 1700 earthquake based on estimates of size, location, displacement and coastal subsidence 93 [Figures 2a,b,c; Witter et al, 2011]. Thus the modeled run-up height of the 1700 tsunami can be used as a basis to 94 investigate possible impacts along the coast and estuaries of South Beach. 95 Figures 2a,b show the model results of tsunami inundation level and flow speed for South Beach assuming the “L” 96 or large sized earthquake (Mw 9.0) for the A.D. 1700 event [Wei, 2017]. The L model assumes a finite-fault source 97 with maximum vertical coseismic displacement of 15.2 m and subsidence of ~1.03 m at South Beach [Figure 2c; 98 Witter et al., 2011]. The Witter et al. (2011) coseismic subsidence estimate differs slightly from the Satake et al (2003)

3

https://doi.org/10.5194/nhess-2020-427 Preprint. Discussion started: 27 January 2021 c Author(s) 2021. CC BY 4.0 License.

99 because it is based on coseismic slip from turbidite records [Goldfinger, 2011], and includes a rupture model with 100 slip partition into a splay fault in the accretionary wedge. The earthquake source duration was not taken into account 101 in the model. 102 103 Two models were used to compute the tsunami inundation levels and flow speed [Wei, 2017] based on four-level 104 telescoped model grids at the spatial resolutions of 1 arc min (~ 1.8 km), 12 arc sec (~ 360 m), 2 arc sec (~ 60 m), 105 and 1/6 arc sec (~ 5 m). The MOST model [Titov and Gonzalez, 1997] is based on the shallow-water wave equations 106 and uses the estimates of coseismic slip to account for deep-water wave propagation. The MOST model then provides 107 the boundary conditions computed from the level-1 grid (Figure 2a) for a Boussinesq model [Zhou et al., 2011], 108 which takes into account frequency dispersion when computing the nearshore wave-propagation field and onshore 109 tsunami inundation in level-2, 3, and 4 grids (Figure 2b shows the coverage of level-4 grid). The digital-elevation 110 and bathymetric grid of Newport-South Beach in level 4 used in the tsunami inundation models has a spatial resolution 111 of 1/6 arc sec (~5 m). It is derived from the Digital Elevation Model provided by the Oregon Department of 112 and Mineral Industries (DOGAMI). This dataset contains lidar data based on DOGAMI Lidar Data Quadrangles for 113 Toledo South, Newport North, and Newport South. The horizontal datum is WGS 84. The vertical datum is NAVD 114 1988, and it is then converted to Mean Higher High Water (MHHW) level, which is the vertical datum in our tsunami 115 inundation models. MHHW is 2.317 m above the NAVD 1988, and 1.185 m above Mean Sea Level (MSL) at Newport 116 according to the datum information at the National Ocean Service (NOS) tide gauge at South Beach. Typically when 117 performing hazard assessments, Mean High Water (MHW) or MHHW is assumed over the entire duration of tsunami 118 [Wei, 2017], and using MHHW as the vertical datum usually gives a more conservative estimate of the tsunami 119 impact. In the present study, we prefer to use MHHW, instead of the actual tidal level, as our model reference level 120 due to: 1) the uncertainty of the time of the event, which is based on estimates from Japanese records (Satake et al. 121 1996), and could vary over a window of 1-2 hours; 2) the uncertainty of the earthquake/tsunami source; and 3) the 122 uncertainty in the amount of sea level change, which is > 0.5 m over the past 300 years based on a rate of 1.77 mm 123 annual increase. The impact of these uncertainties on the model could overshadow the difference between MHHW 124 and the actual tidal level, and adds an additional level of uncertainty to the model results. A Manning’s coefficient of 125 friction of 0.03 is uniformly applied for both the land and ocean components of the tsunami propagation model. To 126 more realistically estimate the tsunami impact produced by the 1700 event, we removed the two jetties at the entrance 127 of Yaquina Bay from the model DEMs, which leads to greater tsunami inundation levels and impact at South Beach. 128 The tsunami model results discussed hereafter are based on the revised DEMs without the jetties. 129 130 The tsunami inundation model presented here indicates the “L” earthquake, with the co-seismic subsidence taken into 131 account, would produce a tsunami that could inundate South Beach to tsunami water levels up to 17 m (Figure 2a), 132 and inundation depths up to 16 m (Figure 3a,b). The height of the water level at the western section of Mike Miller

4

https://doi.org/10.5194/nhess-2020-427 Preprint. Discussion started: 27 January 2021 c Author(s) 2021. CC BY 4.0 License.

133 Park is generally between 12-15 m, and reduces to between 9-12 m on the eastern side. It is important to note that 134 “tsunami water level” is a term used to describe the elevation reached by seawater measured relative to a stated datum 135 (MHHW herein). Whereas “inundation depth” refers to the local water depth, or height of the tsunami above the 136 ground after taking into account the co-seismic subsidence at a specific location, as shown in Figure 2c. However, 137 there is significant variation in the topography of South Beach, and several areas are predicted to experience a range 138 of inundation depths much less than the 16 m maximum. For example, the model shows the amount of inundation 139 decreases eastward of the beach, and the location of the old-growth Douglas Fir stand at Mike Miller State Park in 140 South Beach may be subjected to a range of inundation depth from negligible to as much as 10 m (Figure 3b). 141 Moreover, the South Beach stand would likely have been subjected to flow velocities between 2-10 m s-1 (Figure 142 2B). These velocities are lower than most of the westward portions of the South Beach Peninsula because the stand 143 is located on topography that can be up to 10 m higher elevation than most of the westward terrain. Nevertheless, it 144 would seem these tsunami current velocities would be high enough to cause significant damage to the South Beach 145 trees, through the large mass and momentum of this volume of sea water, that would be observable in the tree growth. 146 147 Lastly, it is worth noting that the “L” earthquake tsunami model presented here also involves the activation of splay 148 faults in the overriding plate above the subduction zone. Motion on these splay faults introduce a larger co-seismic 149 subsidence along the coastline, and therefore represent a more extreme inundation scenario for the A.D. 1700 event 150 than previous models. Based on the turbidites records reported by Goldfinger et al. (2011), the “L” and larger 151 earthquake scenarios occurred four times in the past 10,000 years, and thus is referred as a 2,500-year event, although 152 the general earthquake size class and associated time interval for an “L” event is estimated to be 800 years by Witter 153 et al. (2011). 154 155 4.0 Impacts of Earthquakes and Tsunamis Inundation on Tree Growth 156 157 4.1 Earthquake induced ring growth disturbance 158 Although there is evidence for only moderate levels of ground shaking in coastal Oregon and Washington following 159 the 1700 earthquake [Obermeier and Dickenson, 2000], ground motion during large earthquakes has been shown to 160 cause significant forest disturbance in other earthquake prone regions. As previously mentioned, these earthquake- 161 induced disturbances are caused by felling or damaging trees, inducing local landslides, or altering the stand’s water 162 access [Jacoby et al, 1997]. Trees that survive these disturbances can have a tree-ring record that shows both sudden 163 growth suppression events, or even sudden positive growth events because of possible reduced competition from 164 nearby damaged trees. Moreover, pulses in tree recruitment may follow a large earthquake as young trees colonize 165 gaps left by damaged over-story individuals [Jacoby et al., 1997]. 166

5

https://doi.org/10.5194/nhess-2020-427 Preprint. Discussion started: 27 January 2021 c Author(s) 2021. CC BY 4.0 License.

167 Trees can respond both directly and indirectly to the effects of large earthquakes. Indirect responses can occur due to 168 coseismic environmental changes. For example, Fuller (1912) noted trees died from flooding during the 1811-1812 169 New Madrid earthquakes. Wallace and LaMarche (1979) found coast redwoods (Sequoia sempervirens) and 170 Douglas-firs (Pseudotsuga menziesii) tilted by the 1906 San Andreas Fault earthquake had reaction wood growth 171 starting in 1907. Meisling and Sieh (1980) reported the January 1857 Ft. Tejon earthquake caused conifers to lose 172 their crowns, which reduced ring widths that took many years to return to pre-earthquake growth rates. Jacoby and 173 Ulan (1983) showed the September 1899 Alaska earthquake caused near shore Sitka spruces (Picea sitchensis) to 174 increase growth because coseismic uplift moved the shoreline, resulting in less exposure to wind, salt spray, and root- 175 zone erosion. As for direct responses to earthquake impacts, Jacoby et al., (1988) analyzed conifer tree-ring samples 176 near the epicenter of the 1812 San Juan Capistrano earthquake. A total of nine on-fault trees showed drastic growth 177 reductions in 1813, requiring decades to return to pre-disturbance growth rates. Similarly, Shephard and Jacoby 178 (1989) showed that the 1964 Alaskan earthquake, which caused ~4 m of coseismic uplift, initially induced growth 179 reduction in Sitka Spruces, but the trees eventually responded with wide reaction wood rings in the following years 180 to regain upright positions. 181 182 4.2 Tsunami induced tree-ring growth disturbance 183 Just as trees located near epicenters of large earthquakes can be disturbed and show growth changes from intense 184 shaking and ground displacement, inundation of a coastal forest by a large tsunami should also have a significant 185 impact. Catastrophic tsunami inundation events along the Sumatra and Japanese coasts showed tsunamis have a 186 devastating effect on coastal forests, damaging trees and severely eroded and alter the beach and estuary 187 geomorphology [e.g. Kathiresan and Rajendran, 2005; Udo et al., 2012; Lopez Caceres et al., 2018]. It is expected 188 that inundation by a tsunami would cause significant ring-growth reduction due to physical impact from the wave, 189 prolonged exposure to salt-water, and from tsunami debris that would also physically impact the tree. 190 191 In the U.S. Pacific Northwest, when the A.D. 1700 co-seismic tree-ring growth disturbance is considered, it is largely 192 of trees killed by inundation attributed to co-seismic subsidence [e.g. Atwater and Yamaguchi, 1991]. However, 193 Jacoby et al., [1997] were able to find trees that pre-dated the 1700 Cascadia earthquake and survived subsidence and 194 inundation, which is analogous to the tree-ring growth scenario we observed in South Beach, Oregon. 195 196 Jacoby et al. [1997] collected cores from 33 living Sitka spruce trees that were established earlier than 1700 (i.e. at 197 least 300 years old) that stand along the western Columbia River between Washington and Oregon (~220 km north 198 of South Beach). While 15 of these trees show some evidence of disturbance at 1700, 5 trees showed no disturbance 199 and the remaining 14 could be in either category. There were both unusual decreases and increases in ring-width in 200 disturbed trees. Disturbed trees also showed water-logging and increasing numbers of traumatic resin canals at 1700

6

https://doi.org/10.5194/nhess-2020-427 Preprint. Discussion started: 27 January 2021 c Author(s) 2021. CC BY 4.0 License.

201 (sap-conducting tubes formed by altered cells), but only 2 show reaction wood formation in response to co-seismic 202 tilting or flooding. All trees also display a wide range in the years that ring-width changes begin, with clear decline 203 in growth occurring as early as 1698 and as late as 1702 up to 1706. 204 205 Thus the exact timing of tree-ring disturbances due to an earthquake and the resulting ground motion, coastal land 206 subsidence and tsunami inundation can vary within a few years around the event date. This is because tree growth 207 can be affected by several climatological/meteorological factors, including droughts, cold/heat stress, fires, and 208 windstorms and even insect infestations. However, comparison of coastal growth rings with other regional sites can 209 be used to control for these climate/weather disturbance impacts. Thus, despite this temporal variability, Jacoby et al 210 (1997) conclude the subduction earthquake/subsidence event occurred between the growing seasons of 1699 and 211 1700. Thus it seems likely a combination of the effects from earthquake ground motion, coastal land subsidence, and 212 rapid inundation by several meters of fast moving sea water can be observed in the variation of ring growth 213 chronologies from trees within the impact zone. 214 215 5.0 Tree Ring Growth Chronologies from South Beach, Oregon 216 217 In an attempt to further quantify the widespread effects of the A.D. 1700 earthquake, we obtained tree-ring records 218 from a stand of old-growth Douglas fir trees (Pseudotsuga menziesii) whose ages pre-date 1700 (Figure 4a,b). The 219 stand is located in an Oregon State Park in South Beach, Oregon, roughly 600 m east of Highway 101 (Figure 3a,b). 220 Old growth trees of 300+ years of age are rare along the Oregon coast, thus this stand of trees within the inundation 221 zone presented a unique opportunity to search for direct physical evidence of the impact of a Cascadia Subduction 222 zone earthquake and tsunami inundation in a populated area where tsunami models indicate significant inundation 223 levels and run-up heights. 224 225 To ground-truth the model of the A.D. 1700 tsunami, we collected tree cores at breast height from 37 dominant or 226 codominant old-growth trees at the South Beach site using a 32” increment borer. Two cores were collected from 227 each tree, after which cores were mounted, sanded with increasingly fine lapping film, and cross-dated (Phipps 1985). 228 Each core was then measured using a Velmex TA Tree-Ring Measuring device to the nearest 0.001mm (Velmex, Inc. 229 Bloomfield, NY). Cross-dating was then statistically verified using the program COFECHA [Holmes 1983]. A master 230 growth-increment chronology was then developed by detrending each measurement time series using a negative 231 exponential or regression functions to retain as much low-frequency variability as possible as well as a second 232 chronology developed using 50-year 50% frequency-cutoff cubic spline to highlight interdecadal to interannual 233 growth variability. All chronology construction was performed using the program ARSTAN Cook and Krusic 2005]. 234 Figure 3a,b shows the location of the Douglas Fir trees sampled for this study in relation to the modeled tsunami

7

https://doi.org/10.5194/nhess-2020-427 Preprint. Discussion started: 27 January 2021 c Author(s) 2021. CC BY 4.0 License.

235 run-up heights for South Beach. Of all the trees sampled, a total of twelves cores from eight trees pre-dated 1700 236 (Figure 4b). 237 238 As noted, the tsunami inundation model presented here (Figure 2a,b) indicates the “L” earthquake would produce a 239 tsunami that could inundate the lowlands of South Beach to inundation depths up to 18 m. However, the Douglas fir 240 old-growth stand that is the subject of this study lies on, and along the western edge, of two parallel north-south 241 striking topographic highs (likely paleo-dune ridge lines). The tsunami model presented here indicates that while 242 many of the trees in this area may have experienced as much as ~10 m of inundation depth, several trees are also on 243 high ground and may have experienced much less, or even zero, inundation. 244 245 Tree-ring data detrended using negative exponential functions did not reveal major stand-wide releases or 246 suppressions around 1700 (data not shown) nor did data detrended using the 50-year spline functions. Detailed 247 examination of the growth-ring samples indicates that although individual cores have below-average growth, and one 248 experiences what could be interpreted as a post-1700 growth release, variability around 1700 is not necessarily 249 exceptional in the longer-term context of the ~310 year history of the dataset (Figure 4a,b) . Indeed, there are at least 250 5 other growth reductions in the record that are the same magnitude or larger than the disturbance at A.D. 1700. Most 251 notably, there are large suppressions observed beginning around 1691 and again in 1739 and 1745 (Figure 4a). The 252 1739 reduction has been observed in other old-growth stand chronologies throughout Cascadia and may be due to a 253 significant climatological event such as a drought [Carroll et al. 2005, 2014]. 254 255 5.1 Control Sites in Western Oregon Cascades and Coast Range 256 257 To better detect unusual growth anomalies around 1700, we compared the South Beach Douglas fir tree-ring data to 258 two other Douglas-fir data sets from the Oregon Coast Range and western Cascade Mountains, which would have 259 experienced similar climate conditions but not tsunami inundation. The first of these was an old-growth stand on 260 Marys Peak (~46 km east of South Beach) in the central Coast Range and Browder Creek located ~160 km east in the 261 western Cascade Mountains [Black et al. 2015; Figure 5a,b]. The other was a chronology generated from dead- 262 sampled trees in lakes in the western and central Oregon Coast Range. These lakes had formed when landslides 263 impounded streams, and the preserved drowned trees were then used to establish the date of lake formation, and thus 264 the event [Struble et al. 2020]. Eight lakes had a combined number of 15 trees (and 31 sets of measurements) 265 that pre-dated 1700 and were used to generate a second “control” chronology. As with Mike Miller trees, all 266 measurement time series in the control datasets were detrended with 50-year splines. 267

8

https://doi.org/10.5194/nhess-2020-427 Preprint. Discussion started: 27 January 2021 c Author(s) 2021. CC BY 4.0 License.

268 When compared to the control sites, South Beach tree growth is significantly lower than the lakes or Marys Peak in 269 1700 (t-test of p < 0.001). Moreover, the lakes trees and Marys Peak trees do not significantly differ in growth in 1700, 270 suggesting the South Beach Douglas fir growth is unusually low across sites with similar climatic histories and 271 sensitivities. 272 273 6.0 Discussion 274 275 Analysis of tree-ring data presented here indicates there is a reduction in Douglas-fir tree growth at a site associated 276 with the 1700 Cascadia Subduction Zone earthquake and tsunami in South Beach Oregon relative to other inland, 277 sites in the Oregon coast range. The growth reduction is not outside the range of variability, as illustrated by other 278 much more severe suppressions over the 310 year South Beach chronology (Figure 5a,b). However, there is at least 279 a subtle growth reduction that deviates from other nearby locations. 280 281 The tsunami inundation model presented here, which assumes the “L” or large sized earthquake (Mw 9.0) for the 282 1700 event [Wei, 2017] shows the resulting tsunami would have inundated the South Beach Douglas-fir stand. The 283 growth reduction in Douglas-fir at South Beach, while not absolute conclusive evidence, is at least consistent with 284 such an event and with both the magnitude and multi-year time period of growth reductions observed by Jacoby et al 285 (1997) along the Columbia River ~220 km to the north. 286 287 Although tree rings and various geologic lines of evidence have been useful in establishing the date of the last 288 Cascadia earthquake, there is still some question as to the degree of peak ground motion associated with the event. 289 As previously discussed, a coastal-wide inventory of liquefaction features associated with the 1700 CSZ event 290 indicate only moderate levels of ground shaking in coastal Oregon and Washington [Obermeier and Dickenson, 291 2000]. Mapping seismically induced landslides in the Oregon coast range is potentially another means to assess 292 levels and distribution of seismic shaking impacts from the 1700 CSZ event, given large-magnitude earthquakes in 293 mountainous regions around the world typically trigger thousands of landslides, and slope failures constitute a 294 significant proportion of the damage associated with these events [Stuble et al. 2020]. Recent studies have 295 demonstrated the utility of to date the ages of landslides in these settings [Stuble et al., 2020]. 296 However, despite the Oregon coast range exhibiting thousands of landslides, none have been conclusively associated 297 with the 1700 subduction earthquake, and despite proximity to the megathrust rupture, most deep-seated landslides in 298 the Oregon Coast Range were triggered by rainfall [LaHusan et al., 2020]. Thus a continued search for physical 299 evidence of tsunami inundation, earthquake shaking, and co-seismic landslides is needed to refine expectations of the 300 inundation as well as intensity and distribution of ground shaking during future Cascadia megathrust earthquakes. 301

9

https://doi.org/10.5194/nhess-2020-427 Preprint. Discussion started: 27 January 2021 c Author(s) 2021. CC BY 4.0 License.

302 7.0 Summary 303 We presented a series of tree-ring data from an old-growth Douglas-fir forest in South Beach, Oregon that shows 304 significant growth disturbance at the time of the A.D. 1700 Cascadia subduction zone earthquake. In addition, we 305 presented a new, high resolution model of the 1700 tsunami inundation at South Beach old-growth site. Due to 306 significant variation in the South Beach topography, several areas are predicted to experience water levels up to 17 307 m and a range of inundation depths up to 16 m, however the location of the old-growth stand may be subjected to a 308 range of inundation depths from 0-10 m. To better detect tree growth anomalies near AD 1700, we also compared the 309 South Beach Douglas-fir tree-ring data to two other Douglas-fir data sets from the Oregon Coast Range and western 310 Cascade Mountains, which would have experienced similar climate conditions but not tsunami inundation. When 311 compared to these control sites, South Beach tree growth is significantly lower in 1700, and reaffirms that the South 312 Beach Douglas-fir growth is unusually low for the region. Thus the timing of the observed growth reductions in the 313 South Beach Douglas-fir stand is consistent with these disturbances being associated with the A.D. 1700 Cascadia 314 megathrust earthquake and the resulting tsunami, subsidence and ground motion. Overall, we think our study further 315 supports the view that tree-ring data is a promising tool for providing insights on the spatial distribution of co-seismic 316 impacts from megathrust earthquakes, as well as potential ground-truth information for tsunami inundation models. 317 318 8. Sample Availability 319 Upon acceptance of the manuscript, the Mike Miller, Cape Perpetua, and Marys Peak tree-ring data used in this 320 study will be contributed to the NOAA National Centers for Environmental Information International Tree-Ring 321 Databank, https://www.ncdc.noaa.gov/data-access/paleoclimatology-data/datasets/tree-ring 322 9. Code Availability 323 The tsunami model can be made available with a request to [email protected] followed by a 324 model software training course provided by the NOAA Center for Tsunami Research 325 10. Author Contribution 326 RD prepared the manuscript with contributions from all co-authors. BB and RD collected the tree-ring samples, 327 performed growth disturbance analysis, and wrote the manuscript. YW developed the tsunami model code and 328 performed the simulations, and wrote tsunami section of manuscript. SM drafted initial maps used in manuscript. 329 11. Competing Interests 330 The authors declare that they have no conflict of interest. 331 12. Acknowledgements 332 333 The authors wish to thank the editor and two reviewers. The research in this paper was sponsored by the 334 NOAA/Pacific Marine Environmental Laboratory, PMEL paper contribution number 5184. Yong Wei’s work is 335 funded by the Joint Institute for the Study of the Atmosphere and Ocean (JISAO) under NOAA Cooperative 336 Agreement NA15OAR4320063, Contribution No. 2020-1084. All data is available from the authors upon request, 337 without undue reservation, to any qualified researcher.

10

https://doi.org/10.5194/nhess-2020-427 Preprint. Discussion started: 27 January 2021 c Author(s) 2021. CC BY 4.0 License.

338 References

339 Atwater, B.F. and D.K. Yamaguchi (1991). Sudden, probably coseismic submergence of Holocene trees and grass in coastal 340 Washington State. Geology, 19(7), 706-709. 341 Atwater, B. F., (1992). Geologic evidence for earthquakes during the past 2000 years along the Copalis River, southern 342 coastal Washington, J. Geophys. Res., 97, 1901–1919 343 Atwater, B.F., M. P. Tuttle, E.S. Schweig, C.M. Rubin, D.K. Yamaguchi, and E. Hemphill-Haley (2003). Earthquake 344 recurrence inferred from paleoseismology. Developments in Quaternary Science, 1, 331-350. 345 Atwater, B., M.R. Satoko, K. Satake, T. Yoshinobu, U. Kazue, D.K. Yamaguchi (2006). The orphan tsunami of 1700- 346 Japanese clues t a parent earthquake in the North America. U.S. Geol. Survey Professional Paper 1707, 14p. 347 Black, B.A., Dunham, J.B., Blundon, B.W., Brim-Box, J., Tepley, A.J., 2015. Long-term growth-increment chronologies 348 reveal diverse influences of climate forcing on freshwater and forest biota in the Pacific Northwest. Global Change 349 Biol. 21, 594-604. 350 Carroll, A.L., and Jules, E.S. (2005). Climatic assessment of a 580-year Chamaecyparis Lawsoniana (Port Orford cedar) 351 tree-ring chronology in the Siskiyou mountains, USA. Madrono 52, 114-122, 119. 352 Carroll, A.L., Sillett, S.C., Kramer, R.D. (2014). Millennium-Scale Crossdating and Inter-Annual Climate Sensitivities of 353 Standing California Redwoods. PLOS ONE 9, e102545. doi:10.1371/journal.pone.0102545. 354 Cook, E.R., Krusic, P.J. (2005). ARSTAN v. 41d: A tree-ring standardization program based on detrending and 355 autoregressive time series modeling, with interactive graphics. Tree-Ring Laboratory, Lamont-Doherty Earth 356 Observatory of Columbia University, Palisades, New York, USA. 357 Fuller, M.L. (1912). The New Madrid earthquake, U.S. Geological Survey Bulletin, 494, 119 pp. 358 Goldfinger, C., C.H. Nelson, J.E. Johnson (2003). Holocene earthquake records from the Cascadia Subduction Zone and 359 northern San Andreas Fault based on precise dating of offshore turbidites. Annu. Rev. Earth Planet. Sci., 31, 555- 360 577, doi:10.1146/annurev.earth.31.100901.141246. 361 Goldfinger, C. (2011). Submarine paleoseismology based on turbidite records. Ann Rev. Mar. Sci. v:3, 35-66. 362 Holmes, R.L. (1983). Computer-assisted quality control in tree-ring dating and measurement. Tree-Ring Bull. 43, 69-78. 363 Jacoby, G.C. and L.D. Ulan (1983). Tree ring indications of uplift at Icy Cape, Alaska, related to 1899 earthquakes. J. 364 Geophys. Res., v:88, B11, https://doi.org/10.1029/JB088iB11p09305 365 Jacoby, G.C., P.R. Shepphard, and K.E Sieh (1988). Irregular recurrence of large earthquakes along the San Andreas Fault: 366 Evidence from Trees. Science, 241, 4862, 196-199. 367 Jacoby, G.C., D.E. Bunker, B.E. Benson (1997). Tree-ring evidence for an A.D. 1700 Cascadia earthquake in Washington 368 and northern Oregon. Geology, v:25, 11, 999-1002. 369 Kathiresan, K, and N. Rajendran (2005). Coastal mangrove forest mitigated tsunami. Estuarine, Coastal and Shelf Science, 370 65, 3, 601-606. 371 LaHusen, S. R., A. R. Duvall1, A. M. Booth, A. Grant, B. A. Mishkin, D. R. Montgomery,W. Struble, J. J. Roering, J. Wartman 372 (2020). Rainfall triggers more deep-seated landslides than Cascadia earthquakes in the Oregon Coast Range, USA, 373 Sci Advances, 6 (38), DOI:10.1126/sciadv.aba6790 374 Lopez Caceres, M.L., S. Kakano, J.P. Ferrio, M. Hayashi, T. Nakatsuka, T. Yamanaka, Y. Nobori (2018). Evaluation of the 375 effect of the 2011 tsunami on coastal forests by means of multiple isotopic analyses of tree-rings, Iso. Env. 376 Health Stud., 54:5, 494-507, doi:10.1080/10256016.2018.1495203. 377 Meisling, K.E., and K.E. Sieh (1980). Disturbance of trees by the 1857 Fort Tejon earthquake, CA, J. Geophys. Res., 85, 378 3225-3238. 379 Nelson, A. R., B. F. Atwater, P. T. Bobrowsky, L. A. Bradley, J. J. Clague, G. A. Carver, M. K. Darienzo, W. C. Grant, H. 380 W. Kruger, R. Sparks, T. W. Stafford, and M. Stuiver (1995). Radiocarbon evidence for extensive plate-boundary 381 rupture about 300 years ago at the Cascadia subduction zone, Nature 378, 371–374. 382 Obermeier, S.F., and S.E. Dickenson (2000). Liquefaction evidence for the strength of ground motion resulting from Late 383 Holocene Cascadia Subduction earthquakes, with emphasis on the event of 1700 A.D., Bull. Seism. Soc. Am., 90, 384 4, 876-896. 385 Phipps, R.L. (1985). Collecting, preparing, crossdating, and measuring increment cores. US Geological Survey Water- 386 Resources Investigations Report 85-4148, 1-47.

11

https://doi.org/10.5194/nhess-2020-427 Preprint. Discussion started: 27 January 2021 c Author(s) 2021. CC BY 4.0 License.

387 Priest, G.R., R.C. Witter, Y.J. Zhang, K. Wang (2013). Tsunami inundation scenarios for Oregon, Open-file Report 0-13-19, 388 State of Oregon Department of Geology and Mineral Industries, 14pp. 389 Satake, K., K. Shimazaki, Y. Shuji, and K. Ueda (1996), Time and size of a giant earthquake in Cascadia inferred from 390 Japanese tsunami records of January 1700, Nature, 379, 246-249. 391 Satake, K., K. Wang, and B. F. Atwater (2003), Fault slip and seismic moment of the 1700 Cascadia earthquake inferred 392 from Japanese tsunami descriptions, J. Geophys. Res., 108, 2535, doi:10.1029/2003JB002521, B11. 393 Sheppard, P.R. and G.C. Jacoby (1989). Application of tree-ring analysis to paleoseismology: Two case studies. Geology, 394 17, 226-229. 395 396 Struble, W.T., Roering, J.J., Black, B.A., Burns, W.J., Calhoun, N., Wetherell, L. (2020). Dendrochronological dating of 397 landslides in western Oregon: Searching for signals of the Cascadia A.D. 1700 earthquake. Geol Soc. Am 398 Bulletin, 132, 1775-1791. doi:10.1130/b35269. 399 Titov, V.V. and F.I. Gonzalez (1997). Implementation and testing of the method of splitting tsunami (MOST) model. NOAA 400 Technical Memorandum ERL PMEL-112, 11pp. 401 Udo, K., D. Sugawara, H. Tanaka, K. Imai, and A Mano (2012). Impact of the 2011 Tohoku earthquake and tsunami on 402 beach morphology along the northern Sendai coast. 54, doi.org/10.1142/s057856341250009x. 403 WDNR (2012). Modeling a magnitude 9.0 earthquake on the Cascadia Subduction Zone off the Pacific Coast, 404 http://www.dnr.wa.gov/Publications/ger_seismic_scenario_cascadia.pdf 405 Wallace, R.E. and V.C. LaMarche (1979). Trees as indicators of past movements on the San Andreas Fault, Earthquake 406 Information Bulletin, 2, 127-131. 407 Wei, Y. (2017), Tsunami inundation modeling for the OSU Marine Studies Initiative Building, report submitted to YGH 408 Architecture, p39. 409 Witter, R.C., Y. Zhang, K. Wang, G.R. Priest, C. Goldfinger, L.L. Stimely, J.T. English and P.A. Ferro (2011), Simulating 410 tsunami inundation at Bandon, Coos County, Oregon, using hypothetical Cascadia and Alaska earthquake scenarios, 411 Special Paper 43, State of Oregon Department of Geology and Mineral Industries, 57pp. 412 Yamaguchi, D.K. (1991). A simple method for cross-dating increment cores from living trees. Canadian J. of Forest Res, 413 21(3): 414-416, https://doi.org/10.1139/x91-053. 414 Yamaguchi, D.K., B.F. Atwater, D.E. Bunker, B.E. Benson, M.S. Reid (1997). Tree-ring dating the 1700 Cascadia 415 earthquake. Nature, 922-923, doi:10.1038/40048. 416 Yeats, R.S. (2004). Living with earthquakes in the Pacific Northwest, a survivor’s guide, second edition: Corvallis, Oregon 417 State University Press, 400 p. 418 Zhou, H., C.W. Moore, Y. Wei, and V.V. Titov (2011). A nested-grid Boussinesq-type approach to modelling dispersive 419 propagation and runup of landslide-generated tsunamis. Natural Hazards and Earth System Sciences, 11, 2677- 420 2697 421

12

https://doi.org/10.5194/nhess-2020-427 Preprint. Discussion started: 27 January 2021 c Author(s) 2021. CC BY 4.0 License.

422

423 424 Figure 1: Map showing location of Newport and South Beach along the central Oregon coast. White dot shows position of 425 Mike Miller State Park in South Beach, which is location of Douglas fir tree (Pseudotsuga menziesii) old growth stand whose 426 ages extend back past AD 1700. The state park is located ~ 2 km south of the Newport-Yaquina Bay, ~1.2 km east of the 427 shoreline and ~600 m east of Highway 101. Maps were created using digital elevation data points complied by National Center 428 for Environmental Information (http://ncei.noaa.gov), and State of Oregon’s Department of Geology and Mineral Industries 429 (https://www.oregongeology.org/lidar/).

430 431

13

https://doi.org/10.5194/nhess-2020-427 Preprint. Discussion started: 27 January 2021 c Author(s) 2021. CC BY 4.0 License.

432

433

434 435 Figure 2: (a,b) Model of tsunami inundation level and flow speed for South Beach assuming the “L” or large sized earthquake 436 (Mw 9.0) for the A.D. 1700 event [Wei, 2017]. The L model assumes a finite-area fault-source with maximum coseismic 437 displacement of 15.2 m and subsidence of ~1.03 m at South Beach Contour levels are shown. (c) The L model assumes a 438 finite-area fault-source with maximum coseismic displacement of 15.2 m and subsidence of ~1.03 m at South Beach [Witter 439 et al., 2011]. 440 441

14

https://doi.org/10.5194/nhess-2020-427 Preprint. Discussion started: 27 January 2021 c Author(s) 2021. CC BY 4.0 License.

442 443 444 Figure 3: (a) Model of maximum tsunami inundation depth at South Beach for the “L” sized earthquake A.D. 1700 event; (b) 445 Zoom in view of the tsunami inundation depth at Mike Miller State Park. Gray dotted circles show location of trees used in 446 this study on north side of the Stand. Red numbers are the tree locations whose growth chronologies are shown in Figure 4a,b. 447 White numbers shows trees that were cored, but chronologies do not include the years before AD 1700. Colors on map show 448 inundation depth from the model, implying 0-10 m of inundation depth at the Mike Miller Park Douglas fir stand. Green areas 449 are high ground locations that show no inundation. Yellow line shows, for comparison, the model of maximum run-up height 450 for the Mw 9.2 “XXLarge” earthquake scenario [Priest et al., 2013]. Base maps made using © Google Earth 2016. 451 452

15

https://doi.org/10.5194/nhess-2020-427 Preprint. Discussion started: 27 January 2021 c Author(s) 2021. CC BY 4.0 License.

453 454 455 Figure 4: a) Tree-ring growth records of old-growth Douglas fir trees (Pseudotsuga menziesii) located in Mike Miller State 456 Park, South Beach, Oregon (see Figure 1). Vertical axis shows ring growth in cm, time range covers several decades before 457 and after AD 1700. The color of each growth record was relates to alpha-numeric labels of individual trees shown in legend, 458 with location of trees shown in Figure 3b. Designation “A/B” represents two cores from same tree. Black arrow marks AD 459 1700 date, red arrows highlight the AD 1691, 1738 and 1745 large growth reductions that may have been caused by a 460 significant climatological events. (b) Shows detailed growth record of trees in (a) 4 years before and after 1700 (black arrow). 461

16

https://doi.org/10.5194/nhess-2020-427 Preprint. Discussion started: 27 January 2021 c Author(s) 2021. CC BY 4.0 License.

462 463 464 Figure 5. a) Difference in growth chronologies between Mike Miller and reference sites at Marys Peak, Browder Creek, and 465 Oregon Coast Range lake sites. The A.D. 1700 chronology indicated by red arrow, the significant growth differences between 466 coast and inland sites in 1739-1741 and 1745-1748 are highlighted by red and black lines, respectively. b) Shows location map 467 with Marys Peak and lake sites relative to South Beach. The number of cores available at each site during the 1700 time period 468 are South Beach (14), Marys Peak (28), Coast Range Lakes (31) and Browder Creek (30). The Oregon Coast Range lakes 469 include: Beaver Dam Lake, Carlton Lake, Esmond Lake, Hamar Lake, Kauppi Lake, Klickitat Lake, Soup Lake, and Wasson 470 Lake. Map created using digital elevation data points complied by National Center for Environmental Information 471 (http://ncei.noaa.gov) 472

17